24 March 2020
Towards better Bayesianism; and, underreporting
To date, my updates have been dipping into the flow of information to try to capture high-quality signals. I don’t know about you, but I can’t keep up anymore; there’s just too much information coming at me at once. Plus, all too often I struggle to know how to integrate even very high-quality information with what I already thought I knew and believed, and therefore how to update my expectations.
For my own benefit, at least, I’m going to try a slightly different approach for the next few days to see if it’s more satisfying. I want to try to document what our best / most reliable information is on key topics (and how we know it, and how reliable we think it is — it’s so-called “epistemic status — as well as what critical questions remain.
The idea is to create these as a resource that builds cumulatively — and, it goes without saying, with a lot of help from friends — and then have future updates be with reference to that baseline understanding.
I reserve the right to modify the format as we learn more about what works, but the initial thought is a set of individual posts/pages on key topics / questions, updated as appropriate.
As a first effort, I’ve published one today: What do we know about the degree of underreporting of COVID-19 cases? I’ve labeled it as a “Draft” as it’s a first effort, and I’d very much like to get feedback and suggestions before removing that label.
COVID-19 from the point of view of pulmonologists
This document apparently represents the current views of the “International pulmonologists’ consensus group on COVID-19.” Much of it is beyond me, but it’s fascinating to see how doctors are coming together to share knowledge and best practices.
I found this summary of what is known about why the disease affects children less significant than adults very interesting (and consistent with what I’ve read elsewhere):
Don’t try this at home
There’s been a lot of discussion about the potential of chloroquine, starting with President Trump’s tweet. Here’s why you shouldn’t try to treat yourself.
How best to support COVID-19 efforts?
Recommendations from my friend William MacAskill, Professor of Philosophy at founder of the Effective Altruism movement. Will has an extremely high bar about both the potential impact of projects, and about the extent to which projects represent “value for money” from a donor’s point of view, in terms of impact per dollar deployed.
I wasn’t surprised to see the Johns Hopkins Centre for Health Security on the list, but all of the others were new ones to me.